Article Text

Cholecystokinin infusion: assessing a rather provocative test
  1. Department of Gastrointestinal Surgery,
  2. Imperial College School of Medicine
  3. at the Hammersmith Hospital
  4. Du Cane Road, London W12 ONN, UK
  1. Professor RCN Williamson.

Statistics from

See article on page 571

All gastroenterologists, whether physicians or surgeons, will recognise a group of patients who have typical gall bladder pain but in whom imaging of the biliary tree remains obstinately negative. Incalculable numbers of acalculous patients, many of them middle-aged women, end up having a cholecystectomy almost by default. The results are surprisingly successful, perhaps owing to the powerful placebo effect of a surgical operation. Thus “blind” cholecystectomy relieved the pain in up to 70% of such patients in two historical series as opposed to the 80–85% that could be anticipated in those with symptomatic gallstones.1 ,2 The source of pain in such patients may lie within the biliary tract even if it cannot be clearly recognised. Chronic acalculous cholecystitis is the commonest suggested cause, but its definition varies between reports and is often not stated. Hyperplastic cholecytoses such as cholesterosis and adenomyosis (diverticulosis) may cause symptoms that are improved by cholecystectomy even in the absence of concomitant stones.3 The so called “cystic duct syndrome” is thought to result from postinflammatory stenosis or congenital tortuosity of the cystic duct and might cause right hypochondrical pain by inhibiting normal gall bladder emptying. In those without any clear pathological diagnosis, there could be a painful dysmotility due to asynchronous contraction of the neck and fundus of the gall bladder. All of these conditions would logically be improved by resection of the gall bladder, whereas common extrabiliary causes of pain such as irritable bowel syndrome should not be improved.

As preoperative diagnosis of acalculous disease of the gall bladder is difficult and the response to cholecystectomy is unpredictable, the cholecystokinin (CCK) provocation test has been advocated as a means of identifying patients who will benefit from the operation. There are two components to the test, but not all investigations have clarified the exact criteria for a positive result. The habitual pain should be reproduced during stimulation of the gall bladder muscle by an intravenous infusion of CCK, while measurement of gall bladder volume (for up to one hour) should demonstrate delayed emptying. Many authors have questioned the specificity and reproducibility of the pain response.4 CCK causes pain in some healthy individuals especially if the injection is rapid. It stimulates the smooth muscle of the small and large intestine and the sphincter of Oddi as well as the gall bladder, and these effects could lead to false positive results in those with intestinal motility disorders or sphincter dyskinesia. Ideally CCK studies should be placebo controlled and double blind, with the imaging and histopathology performed by independent observers.

In this issue (see page 571) Smythe and colleagues have investigated 58 patients with acalculous biliary pain by means of a CCK provocation test using serial ultrasound scans to determine the ejection fraction of the gall bladder. This study had the advantages of being prospective and of comparing the symptomatic response to operative and non-operative treatment in both CCK positive and CCK negative patients. After cholecystectomy, 67% of those with a positive test became symptom-free as opposed to 43% of those with a negative test; this difference was not statistically significant. With conservative management, 40% of those with a positive test lost their symptoms as opposed to 75% of those with a negative test, but numbers were small. When the test was repeated postoperatively, 20 of 25 patients converted from positive to negative, but interestingly the five who remained positive had persisting pain. Although the percentage reduction in gall bladder volume within an hour of CCK infusion was lower in positive than negative responders (33 v 52%), this result did not correlate with the symptomatic benefit after cholecystectomy.

The authors conclude that cholecystectomy has a good chance of curing episodic right hypochondrical pain but that the CCK provocation test is of no predictive value. Their study can be criticised on four counts. Firstly, better prediction of symptomatic outcome could have been obtained if the operation had been performed on equal proportions of patients with positive and negative tests; the disproportion in the present study (84 v 54%) will have introduced a verification bias to the statistical analysis. Secondly, although most of the resected gall bladders showed evidence of chronic cholecystitis on histological examination, this can be rather a subjective diagnosis and the criteria are not stated. It is not clear whether the pathologist examined the cystic duct for narrowing or identified any cases of cholecystosis. Thirdly, six months might be too early to assess postoperative pain relief. Fourthly, ultrasonography (or oral cholecystography) may not be the best way of measuring gall bladder emptying as the images are two dimensional and subject to considerable observer bias; a wide variation in ejection fractions (between 10 and 90%) can be found among control subjects.5 CCK scintigraphy is probably more accurate, being three dimensional, computer based and operator independent. A maximal ejection fraction below 35% correlates closely with the histological changes in the resected gall bladders (or cystic ducts) of those with acalculous biliary pain, and it can predict the symptomatic response to operation in such patients.6 ,7

The disappointing results of the present study on the role of the CCK provocation test support those of a previous study of 48 patients in which a positive clinical or radiological response to CCK could not predict the response to cholecystectomy.8 By contrast, a much larger study from Rhodes and colleagues showed that a positive CCK test had an excellent chance of predicting symptomatic benefit. Of 90 patients with pain after CCK infusion, 67% had complete resolution of symptoms postoperatively and another 24% had a noticeable improvement in symptoms.9

In conclusion, a case can still be made for performing CCK provocation studies in patients with biliary pain but a normal gall bladder on conventional imaging. At least two in three of those with a positive test will be helped by cholecystectomy. Although a negative test does not rule out the chance of a good postoperative result, it might encourage a further search for alternative diagnoses. Should that fail, patients may be offered an operation on the understanding that they have only a 50:50 chance of symptomatic relief. Pragmatists may feel that the limitations of the test are such that any fit patient with persisting symptoms should be advised to undergo cholecystectomy, particularly nowadays when minimal access techniques have reduced the metabolic “insult” of this procedure.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles